When large defects occur, bone regeneration can be supported by bone grafting and biophysical stimuli like electric and magnetic stimulation (EMS). Clinically established EMS modes are external coils and surgical implants like an electroinductive screw system, which combines a magnetic and electric field, e.g., for the treatment of avascular bone necrosis or pseudarthrosis. For optimization of this implant system, an in vitro test setup was designed to investigate effects of EMS on human osteoblasts on different 3D scaffolds (based on calcium phosphate and collagen). Prior to the cell experiments, numerical simulations of the setup, as well as experimental validation, via measurements of the electric parameters induced by EMS were conducted. Human osteoblasts (3 × 10(5) cells) were seeded onto the scaffolds and cultivated. After 24 h, screw implants (Stryker ASNIS III s-series) were centered in the scaffolds, and EMS was applied (3 × 45 min per day at 20 Hz) for 3 days. Cell viability and collagen type 1 (Col1) synthesis were determined subsequently. Numerical simulation and validation showed an adequate distribution of the electric field within the scaffolds. Experimental measurements of the electric potential revealed only minimal deviation from the simulation. Cell response to stimulation varied with scaffold material and mode of stimulation. EMS-stimulated cells exhibited a significant decrease of metabolic activity in particular on collagen scaffolds. In contrast, the Col1/metabolic activity ratio was significantly increased on collagen and non-sintered calcium phosphate scaffolds after 3 days. Exclusive magnetic stimulation showed similar but nonsignificant tendencies in metabolic activity and Col1 synthesis. The cell tests demonstrate that the new test setup is a valuable tool for in vitro testing and parameter optimization of the clinically used electroinductive screw system. It combines magnetic and electric stimulation, allowing in vitro investigations of its influence on human osteoblasts.

Automatic reading for water meter is one of the practical demands in smart city applications. Due to the high cost, it is not feasible to replace the old mechanical water meter with a new embedded electronic device. Recently, image recognition based meter reading methods have become research hotspots. However, illumination, occlusion, energy and computational consuming in IoT environment bring challenges to these methods. In this paper, we design and implement a smart water meter reading system to handle this issue. Specifically, we first propose a novel light-weight spliced convolution network to recognize the meter number, which simplifies standard 3 × 3 convolutions by splicing a certain number of 1 × 1 and 3 × 3 size kernel. We then prove the superiority of our network by theoretical analysis. Second, we have implemented the prototype which can handle huge real-time data base on the distributed cloud platform. Base on this system, our system can provide industrial service. Finally, we conduct real-world dataset to verify the performance of the system. The experimental results demonstrate that our proposed light-weight spliced convolution network can reduce nearly 10× computational consuming, 7× model space, and save 3× running time comparing with standard convolution network.INDEX TERMS Water meter reading, spliced convolution network, cyber-physical-system, smart city.

Deep learning has proved to be very effective in video action recognition. Video violence recognition attempts to learn the human multi-dynamic behaviors in more complex scenarios. In this work, we develop a method for video violence recognition from a new perspective of skeleton points. Unlike the previous works, we first formulate 3D skeleton point clouds from human skeleton sequences extracted from videos and then perform interaction learning on these 3D skeleton point clouds. Specifically, we propose two types of Skeleton Points Interaction Learning (SPIL) strategies: (i) Local-SPIL: by constructing a specific weight distribution strategy between local regional points, Local-SPIL aims to selectively focus on the most relevant parts of them based on their features and spatial-temporal position information. In order to capture diverse types of relation information, a multi-head mechanism is designed to aggregate different features from independent heads to jointly handle different types of relationships between points. (ii) Global-SPIL: to better learn and refine the features of the unordered and unstructured skeleton points, Global-SPIL employs the self-attention layer that operates directly on the sampled points, which can help to make the output more permutation-invariant and well-suited for our task. Extensive experimental results validate the effectiveness of our approach and show that our model outperforms the existing networks and achieves new state-of-the-art performance on video violence datasets.

The application of electromagnetic fields to support the bone-healing processes is a therapeutic approach for patients with musculoskeletal disorders. The ASNIS-III s-series screw is a bone stimulation system providing electromagnetic stimulation; however, its influence on human osteoblasts (hOBs) has not been extensively investigated. Therefore, in the present study, the impact of this system on the viability and differentiation of hOBs was examined. We used the ASNIS-III s screw system in terms of a specific experimental test set-up. The ASNIS-III s screw system was used for the application of electromagnetic fields (EMF, 3 mT, 20 Hz) and electromagnetic fields combined with an additional alternating electric field (EMF + EF) (3 mT, 20 Hz, 700 mV). The stimulation of primary hOBs was conducted 3 times per day for 45 min over a period of 72 h. Unstimulated cells served as the controls. Subsequently, the viability, the gene expression of differentiation markers and pro-collagen type 1 synthesis of the stimulated osteoblasts and corresponding controls were investigated. The application of both EMF and EMF + EF using the ASNIS-III s screw system revealed a positive influence on bone cell viability and moderately increased the synthesis of pro-collagen type 1 compared to the unstimulated controls. Stimulation with EMF resulted in a slightly enhanced gene expression of type 1 collagen and osteocalcin; however, stimulation with EMF + EF resulted in a significant increase in alkaline phosphatase (1.4-fold) and osteocalcin (1.6-fold) levels, and a notable increase in the levels of runt-related transcription factor 2 (RUNX-2; 1.54-fold). Our findings demonstrate that stimulation with electromagnetic fields and an additional alternating electric field has a positive influence on hOBs as regards cell viability and the expression of osteoblastic differentiation markers.

No abstract